[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Question about how to proceed with an update breaking another packages tests



On Mon, Jun 06, 2022 at 10:39:00PM +0530, Pirate Praveen wrote:
> On തി, ജൂൺ 6 2022 at 03:11:56 വൈകു +05:30:00 +05:30:00, Nilesh Patra
> <nilesh@debian.org> wrote:
> > >  The problem I am facing is: Updating go-crypto breaks go-pgpmail
> > > [5]. I
> > >  created an issue upstream [6] and
> > >  it seems that it's just an issue with it's
> > >  tests.
> > 
> > I read the github issue. It seems so.
> > But I am not 100% sure if it really is tests - we can only know once we
> > decrypt
> > this in the test code and check back.
> > I am (virtually) maintaining pgp mail since the
> > maintainer is inactive, and pgpmail is used for aerc.
> > It'd be a problem if it is not just the tests.
> > 
> 
> But we can't keep an older version forever too.
> We can't block a required
> update forever. Only a reasonable notice and heads up is how we do
> transitions. Breaking without sufficient notice and time is what we should
> avoid.

Agreed.
But in this case, the latest upstream of pgpmail is failing with
no confirmed ETA, and aerc has a reasonable popcon so I see it as important.
aerc also missed the last debian-stable release due to being unmaintained.

> > 
> > >  - Just ignore the failing tests and update go-crypto
> > 
> > Please never break other's packages.
> 
> Never break without coordinating. Sometimes we have to leave packages behind
> if no one is willing to maintain it. So if you don't care about aerc
> personally, you are not really required to keep maintaining it.

I would do so if I had an option. I want it to be healthy since we do care
about our users.
At this point, I am also central point of contact to upstream regarding updates/maint
in debian.

> You can ask
> for help and even orphan it. [..]

It was in a bitrot for a long time before I stepped in. I can file
an RFH but I find it unlikely for someone to be helping soonish.

> If you don't see a fix in say one or two months, then I
> don't think it is reasonable to block the update.

Let's see. In worst case I can do a patch myself cutting off some sleep.

> Another factor is how urgent the new update is.

It's probably not ultimately urgent.

> In general, we want to update to newer versions
> when possible.

Yeah.

-- 
Best,
Nilesh

Attachment: signature.asc
Description: PGP signature


Reply to: